First of all it's no bother. I was taught me that the price of getting information was giving back when people ask for help.
I say "60 days" because every dog that I've come across responds to the 60 day protocol. I use that number because not everyone can read their dog well, but it results in problems for some. Many dogs will take only 40 days, some 20 days. Some dogs, usually those who will get it in the shorter time periods, will develop a problem if autostims are given for 60 days. They come to believe that the stim is part of the command. (BTW this doesn't not occur with high level stims.) If, at the end of 60 days, you give only the command, not the stim, they think it's NOT the command because part of it is absent. This is rare but it can occur.
The problem is that most people after a few days of getting perfect performance will say, "Let's see how he's doing!" and will try a few commands without the stim. What happens is the same thing that happens with any training where people do this, confusion on the part of the dog. He'll learn in just a few reps of this that sometimes the stim occurs and sometimes it doesn't. If he doing something that he's really interested in, he's liable to "gamble" and see if the stim doesn't come this time. He'll get in the habit of obeying only when it's convenient for him. This is because one of the basic rules of training dogs has been broken; consistency. Performance has not yet become part of the dog's habit but the correction is gone. This is not something that's inherent in the Ecollar, it happens all the time when people try to go off leash before the dog is really ready for it. They find that they get performance for a little while and then it becomes inconsistent.
If you're fairly good at reading your dog you may realize that he's got the idea, that is, he clearly understands that it's his performance that turns off the stim in less than 60 days. But if you guess wrong, you may pay for it down the road.
As far as "before or after the stim . . . " The scientists say that ideally the stim should start 3/10 or 3/100 of a second before the command is given. (I forget which). I'm not good enough to do either one so I give the command and press the button at the same instant. Probably learning would happen faster if the timing of this was perfect, but it still occurs plenty fast enough, as you've no doubt found out.
You can introduce a release command, you may even have one from prior training, but I doubt that at this stage it will have any effect. The dog will stick by your side because he believes that it's a safe spot. No stim occurs there.
I advocate that as soon as you get Velcro dog, you start training the sit which includes the sit at a distance. This will quickly break his "superstition," that near you is a safe spot. He'll come to realize that your command starts the stim and his performance makes it stop.
More than likely the release command will work after the sit training is finished.
Lou Castle has been kicked off this board. He is an OLD SCHOOL DOG TRAINER with little to offer.